Autonomous adaptive data acquisition for scanning hyperspectral imaging

Abstract

Non-invasive and label-free spectral microscopy (spectromicroscopy) techniques can provide quant. biochem. information complementary to genomic sequencing, transcriptomic profiling, and proteomic analyses. However, spectromicroscopy techniques generate high-dimensional data; acquisition of a single spectral image can range from tens of minutes to hours, depending on the desired spatial resoln. and the image size. This substantially limits the timescales of observable transient biol. processes. To address this challenge and move spectromicroscopy towards efficient real-time spatiochem. imaging, we developed a grid-less autonomous adaptive sampling method. Our method substantially decreases image acquisition time while increasing sampling d. in regions of steeper physico-chem. gradients. When implemented with scanning Fourier Transform IR spectromicroscopy expts., this grid-less adaptive sampling approach outperformed std. uniform grid sampling in a two-component chem. model system and in a complex biol. sample, Caenorhabditis elegans. We quant. and qual. assess the efficiency of data acquisition using performance metrics and multivariate IR spectral anal., resp.